1. Field of the Invention
The present invention relates to a white balance adjustment method for a digital image device. More particularly, the present invention relates to a white balance adjustment method for a digital image device capable of improving a white detection capability by using luminance division and view division to reduce errors in the adjustment of the white balance.
2. Description of the Related Art
In general, even when a user photographs the same subject with a digital image camera such as a digital still camera or a digital video camera, colors in the photographs look different according to available lighting conditions such as outdoor sun light, cloudy weather, indoor white-color lamp, fluorescent lamp illumination, and so on. However, people do not feel unnatural so much since human eyes adapt themselves to such light source and accept various shades of white as being a white color. The digital image device fully responds to Red Green Blue (RGB) color components included in light sources each having a different color temperature to reproduce a color having a high color temperature as a greenish-white color and a color having a low color temperature as a reddish-white color. Accordingly, in the above situations, it is necessary to render a mixed white color as solid white, and, even when a color temperature changes, it is preferable for a white balance to be maintained at the color temperature. To accomplish this, the RGB ratio for an achromatic subject is controlled to remain a certain value, or a color difference signal such as Red-Yellow (R-Y) or Blue-Yellow (B-Y) is controlled to remain at zero. This is referred to as white balance adjustment. That is, for a greenish-white color, a red (R) gain is increased and a blue (B) gain is decreased, and, for a reddish-white color, the B gain is increased or the R gain is decreased.
In order to precisely adjust the white balance, it is necessary to precisely detect a reference white color. In order to accomplish this, a white color is ideally set which is used as a reference for a certain image-photographing environment, and the color is used to perform white balancing for a digital image device. However, it is troublesome to do this whenever photographs are taken. A method has been developed to determine a white color based on the characteristics of a subject. This method assumes that a color for a captured picture generally comes close to an achromatic color, that is, to a zero, when an integral value is used with respect to all color components included in the captured picture, which can effectively detect a white color if the captured picture has a wide color distribution inclusive of diverse colors. However, if the method is applied to instances of a narrow color distribution such as a picture photographed at sunset, on red tomatoes, or the like, it can result in an unnatural looking photograph, so means to avoid the unnaturalness need to be taken. That is, a method is required in which the white color is detected with the center portion of the captured picture cut off, portions with low color concentration levels are extracted except for portions with colors highly concentrated on the captured picture, or the like. The proposed method should be a luminance division-based method that divides a luminance level range, inputs color data by divided luminance level range, and detects a white color, a view division-based method that divides a picture-captured view into smaller views, inputs color data by divided view, and detects a white color, or the like.
However, such a luminance division-based method based on luminance levels may have a problem of having high errors when colors are corrected based on the white color data detected since the white color data detected includes solid colors and the like which have high luminance. In addition, the view division-based method may have the problem of having high errors when colors are corrected based on the detected white color since all colors existing on each and every divided view are mixed to detect the white color.